Novel benzoxazole inhibitors of mPGES-1

Article abstract of DOI:10.1016/j.bmcl.2012.10.040

A novel series of potent benzoxazole mPGES-1 inhibitors has been derived from a hit from a high throughput screen. Compound 37 displays mPGES-1 inhibition in an enzyme assay (0.018 μM) and PGE-2 inhibition in a cell-based assay (0.034 μM). It demonstrates 500- and 2500-fold selectivity for mPGES-1 over COX-2 and 6-keto PGF-1α, respectively. In vivo PK studies in dogs demonstrate 55% oral bioavailability and an 7 h half-life.

Full text of DOI:10.1016/j.bmcl.2012.10.040

Bioorganic & Medicinal Chemistry Letters 23 (2013) 907–911
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel benzoxazole inhibitors of mPGES-1
Natasha Kablaoui a, Snahel Patel b, Jay Shao a, Douglas Demian c, Keith Hoffmaster d, Francioise Berlioz e,
Michael L. Vazquez f, William M. Moore g, Richard A. Nugent a,
⇑
a Pfizer Research Technology Center, 620 Memorial Dr., Cambridge, MA 02139, USA
b Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, USA
c Blue Sky Biotech, Inc., 60 Prescott St., Worcester, MA 01605, USA
d Novartis Inc., 250 Massachusetts Ave., Cambridge, MA 02139, USA
e Vertex, 130 Waverly St., Cambridge, MA 02139, USA
f Pfizer Global Research, 445 Eastern Point Road, Groton, CT 06340, USA
g Monsanto Company, 800 N. Lindbergh Blvd., St. Louis, MO 63167, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of potent benzoxazole mPGES-1 inhibitors has been derived from a hit from a high
throughput screen. Compound 37 displays mPGES-1 inhibition in an enzyme assay (0.018 M) and
PGE-2 inhibition in a cell-based assay (0.034 M). It demonstrates 500- and 2500-fold selectivity for
Received 20 June 2012
Revised 2 October 2012
Accepted 8 October 2012
Available online 16 October 2012
l
l
mPGES-1 over COX-2 and 6-keto PGF-1
bioavailability and an 7 h half-life.
a
, respectively. In vivo PK studies in dogs demonstrate 55% oral
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
mPGES-1
Benzoxazole
Selective
For patients with inflammatory pain caused by osteoarthritis
(OA) and rheumatoid arthritis (RA), the first line of treatment is
nonsteroidal anti-inflammatory drugs (NSAIDs) and COX-2 inhibi-
tors (COXIBs). The COXIBs are superior to NSAIDs due to their low-
er incidence of gastrointestinal side effects, but they have shown
renal toxicity and increased cardiovascular risk.1 Alternate mecha-
nisms for treating inflammatory pain may address some of the
shortfalls of the current first line treatments.
up-regulated PGE2 synthesis associated with inflammatory disor-
ders while sparing constitutive PGE2, prostacyclin (PGI2), and
thromboxane (TXA2) production for GI, renal, and cardiovascular
homeostasis.
There are several reported small molecule mPGES-1inhibitors in
the literature (Fig. 1). MK-886, 1,6 an indole-based 5-lipoxygenase-
activating protein (FLAP) inhibitor (IC50 = 0.026–0.1
lM),
displays mPGES-1 inhibition for rat (IC50 = 8.2 M) and human
l
Prostaglandin E2 (PGE2) is an inflammatory mediator that plays
a central role in controlling pain and swelling, and is found in the
synovial fluid of patients with OA and RA.2,3 The biosynthesis of
PGE2 begins with the cleavage of arachidonic acid (AA) from mem-
brane phospholipids by phospholipase A2, followed by the conver-
sion of AA to PGH2 by cyclooxygenase (COX), and finally to the
production of PGE2 by PGE synthase (PGES).4 Three forms of PGES
have been reported: cytosolic PGES (cPGES), microsomal PGES-1
(mPGES-1), and microsomal PGES-2 (mPGES-2).4 Both cPGES and
mPGES-2 are constitutively expressed in a variety of tissues, while
mPGES-1 is up-regulated under inflammatory conditions.2,4,5 Syn-
thesis of excessive PGE2 at inflammatory sites is caused by the up-
regulation of COX-2 and mPGES-1 in response to inflammatory
mediators. An inhibitor of mPGES-1 would be expected to inhibit
(IC50 = 1.6 lM). Merck–Frosst scientists have reported on several
inhibitor classes, including 2,7 which is structurally similar to
MK-886. Compound 2 is a potent and selective mPGES-1 inhibitor
(IC50 = 0.003 lM) on the recombinant human mPGES-1 enzyme,
but a 2000-fold shift was observed for PGE2 inhibition in a IL-1b
stimulated A549 whole cell. More recently Merck–Frosst has dis-
closed MF63 3,8 a non-acidic, phenanthreneimidazole, which bears
no structural resemblance to previous inhibitors, but is highly ac-
tive (IC50 = 0.001
shown to be a selective and powerful inhibitor (IC50 = 0.001
Pfizer has shown that biaryl oxicams, such as 5,11 are selective
inhibitors of mPGES-1 (IC50 = 0.016 M), but are highly protein
l
M).9 The trisubstituted urea 410 has also been
M).
l
l
bound. Herein we report a new series of potent and selective benz-
oxazole mPGES-1 inhibitors, which are efficacious in cellular as-
says and are orally bioavailable in dogs.
Screening of the Pfizer compound collection identified 6 as a
⇑
modest inhibitor of human mPGES-1, with an IC50 = 1.2
(Fig. 2). Examination of other, related benzoxazoles through
lM
Corresponding author.
E-mail address: richardnugent@flatleydiscoverylab.com (R.A. Nugent).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.10.040

908
N. Kablaoui et al. / Bioorg. Med. Chem. Lett. 23 (2013) 907–911
Ph
S
CO2H
Cl
CO2H
Cl
F
N
N
Cl
NC
NC
2
1 (MK-886)
N
N
H
N
3 (MF-63)
Cl
Cl
O
Cl
OH
O
N
H
N
N
H
NH
S
O
O
4
5
Figure 1. Previously reported mPGES-1 inhibitors.
Cl
Cl
HN
O
HN
O
N
O
N
O
N
SMe
N
6
7
Figure 2. Lead compounds from HTS screening.
similarity and substructure searching of the Pfizer collection led to
the discovery of 7, which has increased potency (IC50 = 0.85 M)
and selectivity for mPGES-1 (PGE2) over COX-2 (PGF2 ) in an IL-1
is dissolved in DCM, cooled to 0 °C, then treated with chlorine
gas until the solid dissolves. The reaction is warmed to rt, and stir-
red for 4 h, at which time HPLC shows complete consumption of
starting material. After workup, 10 is isolated in 87% yield. Next,
10 and the ester 11 are dissolved in DCM, treated with triethyl-
amine, and stirred at rt for 2 h. The solvent is evaporated and the
residue dissolved with MeOH/water, followed by hydrolysis of
the ester with NaOH to give 12 in 99% yield. Treatment of 12 with
cyclopentylamine, HBTU, and triethylamine in DMF completes the
synthesis of 7. Overall yields are typically >80%.
Compound 7 is a very modular structure, and the general syn-
thetic scheme allows for a systematic examination of the contribu-
tions of the various motifs to mPGES potency and physical
properties. Tables 1–3 illustrate the effects on mPGES inhibitory
activity of modifications to the piperidine, the amide linker, and
l
a
stimulated fetal fibroblast cell assay12 (PGE2 IC50 = 1.01
IC50 >100 M). Compound 7 offers several positive attributes,
lM; PGF2a
l
including low molecular weight and clogP, which allow significant
changes to the structure without destroying its drug-like character.
Compound 7 is stable to human liver microsomes and is predicted
by the Pampa assay13 to have high permeability; a drawback is that
it displays high protein binding (99.5%), which negatively impacts
its efficacy in whole blood assays.
Analogs were synthesized according to Scheme 1, which details
the synthesis of compound 7. Commercially available 2-amino-
4-chloro phenol and 8 are heated in refluxing pyridine for 1 h;
work up gives crude thiol 9 in 90% yield. Without purification, 9
OH
a, b
O
S
HN
+
+
CO2Et
X
EtO
SK
N
Cl
NH2
Cl
8
9
11
X=SH
10 X= Cl
O
O
O
e
O
N
c, d
N
N
N
HN
Cl
OH
Cl
7
12
Scheme 1. Reagents and conditions: (a) pyridine, reflux, 90%; (b) Cl2 (g), DCM, 87%; (c)Et3N, DCM; (d) MeOH/H2O (9:1), NaOH; (e) cyclopentylamine, HBTU, Et3N, DMF.

N. Kablaoui et al. / Bioorg. Med. Chem. Lett. 23 (2013) 907–911
909
Table 1
Table 2
SAR of mPGES-1 inhibition by varying linker shape
Survey of novel amides
Cl
N
O
Cl
R
O
N
N
R
mPGES-1
IC50 M)
O
(
l
R
HN
7
13
14
15
0.853
N
Compd
R
cLogP
mPGES
IC50 M)
(l
O
7
2.65
2.92
0.853
HN
>10
NH
N
19
>30
N
O
HN
20
21
2.09
3.21
3.01
1.68
>10
>10
N
HN
HN
O
HN
S
N
N
O
O
22
23
3.76
3.16
0.821
0.242
HN
H
N
N
16
12.9
HN
O
H3C
Ph
the amide substituent, while holding the 4-chloro benzoxazole
constant. The modifications to the piperidine and amide linker
are not tolerated, but changes to the amide substituent improve
mPGES potency and physical properties.
HN
HN
24
4.20
0.291
Modifications to the piperidine and amide functionality are
shown in Table 1. Compounds 13–16 are synthesized by treating
10 with the appropriate amine as described above. Simple modifi-
cations to this region are not tolerated. For example, opening the
piperidine ring, 13, or replacing it with a piperazine, 14, results
in loss of all activity. The homopiperizine, 16, has measurable inhi-
bition, but it is very weak. The sulfonamide 15 is also inactive.
The amide substituent is a more fruitful area for improving
mPGES activity through structural modifications. Parallel synthesis
of analogs 19–37 is accomplished as described in Scheme 2. Com-
pound 17 and amines 18a–m are dissolved in DMF and treated
with triethylamine and HBTU and stirred overnight. After workup,
the crude material is treated with 1 M HCl in MeOH, and finally
with 10 and triethylamine in DMF. The resulting material is puri-
fied by reverse phase HPLC to yield the desired compound in
>95% purity.
25
26
27
2.11
2.46
3.90
7.18
8.15
0.826
HO2C
MeO
HN
O
O
HN
HN
28
4.41
0.176
Table 2 details changes in ring size and substitution pattern of
the amide. A dramatic loss of activity is observed when the second-
ary amide of 7 is replaced with the tertiary amide, 19. Changes to
the ring size of the amide substituent does not play a significant
role in activity, as there is only a modest improvement in potency
as the ring size increases from four to seven carbons, 20–22. This is
likely due to increased lipophilicity, as measured by the steady in-
crease of clogP. Derivatization of the cyclopentyl ring, 23–25 also
demonstrates a similar improvement in potency with increased
lipophilicity. Comparison of compounds 23 and 24, with a methyl
and phenyl ring in the 2-position, respectively, shows no difference
in potency despite the much larger phenyl ring in 24; this suggests
that there may be a large, hydrophobic pocket near the amide
nitrogen. Ethers 26–28 have increasingly large substituents in that
2-position, probing this space. While potency for these compounds
still tracks with lipophilicity, compounds 27 and 28 (IC50 = 0.826
and can be modified to provide a wide variety of 2-position
substituents.
The effect of substituent chirality is explored in Table 3. Several
sets of diastereomers of different ring size and lipophilicity are
shown. Two diastereomer sets with low lipophilicity were exam-
ined. Serendipitously, in both the 1,4-substituted cyclohexyl exam-
ple (29/30) and the 1,2-substituted cyclopentyl (31/32), the cis
diastereomers are more active than trans. The other sets in Table 3
are 1,2-substituted cyclopentyl rings with higher lipophilicity
(33/34 and 35/36/37), and no difference in activity between the
diastereomers is apparent; these lipophilic compounds are also
more potent, entering into the low double-digit nanomolar range.
Compound 34 is also very potent in the whole cell PGES fetal fibro-
blast (FF) assay, but inactive in the human whole blood (HWB) as-
say.11 The lack of whole blood activity may be due to the high PPB
(>99%) measured for 34. Compounds 35/36/37 also show a high de-
gree of activity in the enzyme assay and FF assay, although the po-
tency in the FF drops off for the cis isomer, 35. All three compounds
and 0.176 lM, respectively) show that the presumed pocket is
large and flexible. The amines necessary to synthesize these com-
pounds are accessible through a procedure described by BASF14,15

910
N. Kablaoui et al. / Bioorg. Med. Chem. Lett. 23 (2013) 907–911
Table 3
Table 4
Study of clogP and stereochemistry
In vitro PK assessment of 37
Pampa Pe
Rat PPB (%)
HLM T1/2 (min)
RLM T1/2 (min)
Cl
(Â10À6 cm/sec)
O
1.3
98
10.8
7.2
N
N
O
R
Table 5
In vivo PK profile of 37
Dog (PO)a
Compd Structure
cLogP mPGES
IC50 M) IC50
FFb
HWBc
M) IC50 (lM)
Parameter
Dog (iv)
(
l
(
l
Dose (mg/kg)
Cl (mL/mg/kg)
MRTb (h)
0.25
2.1
8.1
1.0
—
7.2
—
—
0.5
—
OH
29
1.74
–
0.383
1.26
11.1
NT
HN
Vss (L/kg)
—
1.3
—
0.5
55.0
Cmax
(l
M)
OH
T1/2 (h)
Tmax (h)
F (%)
30
>19.1
>10.0
HN
OH
a
Data is reported as mean of three animals.
31
2.05
0.785a
3.20a
NT
NT
NT
NT
NT
NT
HN
OH
Table 6
32
—
HN
Assay
IC50
0.018
>10.0
0.035
(l
M)
OPh
Human mPGES-1
Human PGDS
33
4.67
0.0431
0.034
0.082
0.040
0.018
IL-1 stimulated fetal fibroblast (PGE)
IL-1 stimulated fetal fibroblast with SnCl2 (PGF2
Rat mPGES-1
IL-1 stimulated rat synovial fibroblast assay
LPS stimulated human whole blood
LPS stimulated dog whole blood
HN
a)
>100
>10
>30
OPh
0.934
1.23
34
—
0.0847 >33.6
HN
Bn
O
35
4.20
0.618
0.296
76.6
27.4
half-life in both species. However, the results of a dog PK study,
shown in Table 5, belie the in vitro predictions, possibly due to dif-
ferences in plasma protein binding across species. Compound 37
exhibits 55% bioavailability with a 7.2 h half-life. The high clear-
ance predicted from rat or human liver microsomes does not pre-
dict the clearance of 37 observed in vivo in dogs. However, in vitro
half-life in human hepatocytes measures 80 min, which is more in
line with that observed in dogs.
HN
Bn
O
36
—
HN
Bn
O
37
—
0.0347 7.56
Table 6 shows the inhibitory effects of compound 37 on other
prostanoids. Selectivity for PGES over PGDS and PGF2 alpha is
shown through enzymatic and fetal fibroblast assays, respectively.
Compound 37 is not effective against rat mPGES-1, either in an en-
zyme assay or in IL-1 stimulated rat synovial fibroblasts,12 demon-
strating a species dependence despite the high sequence homology
between human and rodent mPGES-1. This discrepancy has also
been observed by Côté et al.8 for a phenanthrene imidazole based
series of mPGES-1 inhibitors. Compound 37 does, however, show
similar mPGES inhibitory activity in LPS stimulated dog and human
whole blood assays.9
HN
a
Assay format slightly different for these compounds, with a lower amount of
DMSO content.
b
c
FF = whole cell PGES fetal fibroblast assay.
HWB = human whole blood assay.11
show some activity in the HWB assay, but 37 is the most potent in
all three assays.
Further profiling of compound 37 is shown in Table 4. The com-
pound is also highly protein bound, 98% in rat plasma, which ex-
plains the significant decrease in potency in the human whole
blood cell assay. It is predicted to show poor permeability and is
highly metabolized in rat and human microsomes with a very short
In summary, we have developed potent mPGES-1 inhibitors
from
a hit in a high throughput screen. These compounds
demonstrate better cell activity over previously reported series.
Cl
O
a,b,c
N
N
+
R NH2
BocN
CO2H
O
17
18 a-m
19-37
R
Scheme 2. Reagents and conditions: (a) HBTU, Et3N, DMF; (b) 1 N HCl, MeOH, 18 h; (c) 10, Et3N, DMF, 50 °C, 18 h.

N. Kablaoui et al. / Bioorg. Med. Chem. Lett. 23 (2013) 907–911
4. Park, J. Y.; Pillinger, M. H.; Abramson, R. B. Clin. Immunol. 2006, 119, 229.
911
Compound 37 displays mPGES-1 inhibition of 0.018 lM and selec-
5. Li, X.; Afif, H.; Cheng, S.; Martel-Pelletier, J.; Pelletier, J. P.; Ranger, P.; Fahmi, H.
J. Rheumatol. 2005, 32, 887.
6. Mancini, J. A.; Blood, K.; Guay, J.; Gordon, R.; Claveau, D.; Chan, C.-C.; Riendeau,
D. J. Biol. Chem. 2001, 276, 4469.
7. Riendeau, D.; Aspiotis, R.; Ethier, D.; Gareau, Y.; Grimm, E. L.; Guay, J.; Guiral, S.;
Juteau, H.; Mancini, J. A.; Méthot, N.; Rubin, J.; Friesen, R. W. Bioorg. Med. Chem.
Lett. 2005, 15, 3352.
8. Côté, B.; Boulet, L.; Brideau, C.; Claveau, D.; Ethier, D.; Frenette, R.; Gagnon, M.;
Giroux, A.; Guay, J.; Guiral, S.; Mancini, J.; Martins, E.; Masse, F.; Methot, N.;
Riendeau, D.; Rubin, J.; Xu, D.; Yu, H.; Ducharme, Y.; Friesen, R. W. Bioorg. Med.
Chem. Lett. 2007, 17, 6816.
tivity over COX-2 in the human fetal fibroblast cell assay. It inhibits
PGE2 production in the LPS/human whole blood assay with an
IC50 ꢀ7
lM, with the discrepancy between enzymatic and whole
blood assays due to the high protein binding of the molecule. Co-
mound 37 is well absorbed in dogs, with an unbound exposure
>3x mPGES-1 IC50 for 6 h following an oral dose.
Acknowledgments
9. Xu, D.; Rowland, S. E.; Clark, P.; Giroux, A.; Côté, B.; Guiral, S.; Salem, M.;
Ducharme, Y.; Friesen, R. W.; Méthot, N.; Mancini, J.; Audoly, L.; Riendeau, D. J.
Pharmacol. Exp. Ther. 2008, 326, 754.
10. Chiasson, J.-F.; Boulet, L.; Brideau, C.; Chau, A.; Claveau, D.; Côté, B.; Ethier, D.;
Giroux, A.; Guay, J.; Guiral, S.; Mancini, J.; Massé, F.; Méthot, N.; Riendeau, D.;
Roy, P.; Rubin, J.; Xu, D.; Yu, H.; Ducharme, Y.; Friesen, R. W. Bioorg. Med. Chem.
Lett. 2011, 21, 1488.
The authors wish to thank Per-Johan Jakobsson of the Karo-
linska Institute for the development of a high throughput assay
allowing the identification of the lead compounds. We also grate-
fully acknowledgment the support and contributions of Douglas
Otte, Michael Pollastri, Joan Kelly, Qing Cao, Jim Gierse, Gina Jer-
ome, Gabriel Mbalaviele, and Alex Shaffer.
11. Wang, J.; Limburg, D.; Carter, J.; Mbalaviele, G.; Gierse, J.; Vazquez, M. Bioorg.
Med. Chem. Lett. 2010, 20, 1604.
12. Mbalaviele, G.; Pauley, A.; Shaffer, A.; Zweifel, B.; Mathialagan, S.; Mnich, S.;
Nemirovskiy, O.; Carter, J.; Gierse, J.; Wang, J.; Vazquez, M.; Moore, W.;
Masferrer, J. Biochem. Pharmacol. 2010, 79, 1445.
References and notes
13. Kansy, M.; Senner, F.; Gubernator, K. J. Med. Chem. 1998, 41, 1007.
14. Siegel, W.; Haderlein, G.; Staeb, T. PCT Int. Appl., WO 2007 074047, 2007; Chem.
Abstr. 147, 143055.
1. Melnikova, I. Nat. Rev. Drug Disc. 2005, 4, 453.
2. Kojima, F.; Kato, S.; Kawai, S. Fundam. Clin. Pharmacol. 2005, 19, 255.
3. Futani, H.; Okayama, A.; Matsui, K.; Kashiwamura, S.; Sasaki, T.; Hada, T.;
Nakanishi, K.; Tateishi, H.; Maruo, S.; Okamura, H. J. Immunother. 2002, 25, S61.
15. Govindaraju, T.; Kumar, V. A.; Ganesh, K. N. J. Org. Chem. 2004, 69, 5725.